Internet path creates different experience in way application works

We have an internet application we use in Southeastern Michigan that
is developed in Maryland. We find that the application often see
idiosyncracies that the developer in Maryland does not see. I
attribute this to the way to application handles the difference in the
internet path somehow. Why should we see an application behave any
differently based on the longer path or different path the data takes?

Not being an internet protocol expert, I can only surmise what is
happening to the data between the application server and our thin
clients on internet explorer. It appears that some data is not being
transmitted, not being acknowledged or possibly arriving in an order
differently than was intended. That seems hard to believe, since I
would expect all the packets to be assembled in the proper order and
that internet protocol would allow for "resends" for data that never
made it.

microsoft.public.windowsxp.network_web

eganders wrote:

> We have an internet application we use in Southeastern Michigan that
> is developed in Maryland. We find that the application often see
> idiosyncracies that the developer in Maryland does not see. I
> attribute this to the way to application handles the difference in the
> internet path somehow. Why should we see an application behave any
> differently based on the longer path or different path the data takes?
>
> Not being an internet protocol expert, I can only surmise what is
> happening to the data between the application server and our thin
> clients on internet explorer. It appears that some data is not being
> transmitted, not being acknowledged or possibly arriving in an order
> differently than was intended. That seems hard to believe, since I
> would expect all the packets to be assembled in the proper order and
> that internet protocol would allow for "resends" for data that never
> made it.
>
> microsoft.public.windowsxp.network_web

If the app on the server is accessed via a browser on the client, then
the differences may be due to using different browsers. Unfortunately,
many lazy designers only test with IE.
--
Cheers, Bob

On Mon, 4 Feb 2008 05:16:36 -0800 (PST), eganders <eganders@yahoo.com> wrote:

>We have an internet application we use in Southeastern Michigan that
>is developed in Maryland. We find that the application often see
>idiosyncracies that the developer in Maryland does not see. I
>attribute this to the way to application handles the difference in the
>internet path somehow. Why should we see an application behave any
>differently based on the longer path or different path the data takes?
>
>Not being an internet protocol expert, I can only surmise what is
>happening to the data between the application server and our thin
>clients on internet explorer. It appears that some data is not being
>transmitted, not being acknowledged or possibly arriving in an order
>differently than was intended. That seems hard to believe, since I
>would expect all the packets to be assembled in the proper order and
>that internet protocol would allow for "resends" for data that never
>made it.
>
>microsoft.public.windowsxp.network_web

Both the DNS servers that the two installations use, and the MTU settings on
each computer, can make a difference here.

DNS:
<http://nitecruzr.blogspot.com/2006/05/identifying-dns-problem-in-your.html>
http://nitecruzr.blogspot.com/2006/0...m-in-your.html

MTU:
<http://nitecruzr.blogspot.com/2007/06/determining-mtu-to-single-server.html>
http://nitecruzr.blogspot.com/2007/0...le-server.html

--
Cheers,
Chuck, MS-MVP 2005-2007 [Windows - Networking]
http://nitecruzr.blogspot.com/
Paranoia is not a problem, when it's a normal response from experience.
My email is AT DOT
actual address pchuck mvps org.